Model: SoLU Model: 6 Layers, 3072 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 2623 in Layer 4

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-6l')

Text #0

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.2120. Min Act: -0.0001

Data Index: 573267 (C4 (Web Text))

Max Activating Token Index: 686

Text #1

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.2042. Min Act: -0.0001

Data Index: 752915 (C4 (Web Text))

Max Activating Token Index: 571

Text #2

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1932. Min Act: -0.0001

Data Index: 1052342 (C4 (Web Text))

Max Activating Token Index: 910

Text #3

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1800. Min Act: -0.0001

Data Index: 629501 (C4 (Web Text))

Max Activating Token Index: 26

Text #4

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1895. Min Act: -0.0000

Data Index: 864227 (C4 (Web Text))

Max Activating Token Index: 801

Text #5

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1815. Min Act: -0.0001

Data Index: 1276157 (C4 (Web Text))

Max Activating Token Index: 340

Text #6

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1810. Min Act: -0.0000

Data Index: 375321 (C4 (Web Text))

Max Activating Token Index: 36

Text #7

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1857. Min Act: -0.0001

Data Index: 758658 (C4 (Web Text))

Max Activating Token Index: 938

Text #8

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1790. Min Act: -0.0001

Data Index: 515903 (C4 (Web Text))

Max Activating Token Index: 87

Text #9

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1757. Min Act: -0.0001

Data Index: 5845 (C4 (Web Text))

Max Activating Token Index: 148

Text #10

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1683. Min Act: -0.0001

Data Index: 1220092 (C4 (Web Text))

Max Activating Token Index: 62

Text #11

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1805. Min Act: -0.0001

Data Index: 961251 (C4 (Web Text))

Max Activating Token Index: 1010

Text #12

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1680. Min Act: -0.0001

Data Index: 812307 (C4 (Web Text))

Max Activating Token Index: 252

Text #13

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1619. Min Act: -0.0001

Data Index: 1056982 (C4 (Web Text))

Max Activating Token Index: 194

Text #14

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1645. Min Act: -0.0001

Data Index: 47749 (C4 (Web Text))

Max Activating Token Index: 966

Text #15

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1719. Min Act: -0.0001

Data Index: 422867 (C4 (Web Text))

Max Activating Token Index: 502

Text #16

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1777. Min Act: -0.0001

Data Index: 955017 (C4 (Web Text))

Max Activating Token Index: 238

Text #17

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1642. Min Act: -0.0001

Data Index: 426151 (C4 (Web Text))

Max Activating Token Index: 477

Text #18

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1682. Min Act: -0.0001

Data Index: 1272165 (C4 (Web Text))

Max Activating Token Index: 484

Text #19

Max Range: 0.2120. Min Range: -0.2120

Max Act: 0.1731. Min Act: -0.0001

Data Index: 527020 (C4 (Web Text))

Max Activating Token Index: 648